The primary focus of this research is to develop new protein engineering methods for the production of novel enzymes. Novel enzymes could perform functions not observed in biology and perform functions in non-biological environments. Novel enzymes could find diverse applications in chemistry, biology, biotechnology and medicine. This research uses an innovative approach that couples computational protein design methods with combinatorial library screening and selection methods to produce large collections of de novo proteins. De novo proteins are proteins engineered completely from scratch in the laboratory, are not derived from natural proteins, and do not have sequence homology with naturally occurring proteins. De novo proteins are attractive as novel enzymes because they are not constrained by the evolutionary history of natural proteins and so they may be more likely to perform non-biological functions or to perform functions in non-biological environments. To achieve the goals of this research, I will computationally design and experimentally produce large libraries of de novo proteins; I will use a high-throughput folding reporter assay to isolate large sub-libraries of well-folded de novo proteins; and I will use screens, selections, and directed evolution to identify and evolve de novo enzymes. The mixed computational and experimental approach used in this research leverages the most powerful feature of computational protein design: the ability to rapidly identify favorable sequence space, with the most powerful feature of library screening and selection methods: the experimental testing of millions of sequences. Libraries generated in this research will be tagged with folding-reporter green fluorescent protein (FR-GFP). Well-folded proteins tagged with FR-GFP have bright fluorescence and poorly folded proteins have low fluorescence. Fluorescence activated cell-sorting can then be used to isolate populations of well-fold proteins based on fluorescence. To identify functional proteins from these libraries, I will screen the ability of lirary proteins to rescue conditionally lethal E. coli gene deletions, auxotrophs. De novo proteins that rescue auxotrophs possess a function that enables the auxotroph to live. To improve weak phenotypes, I will use a novel directed evolution scheme that simultaneous selects for functional and stable enzymes. The computational and experimental methods developed in this research are straightforward to use and highly general and could gain wide acceptance in research. The knowledge gained and methods developed in this research will make it possible to rapidly and reliably engineer novel enzymes for applications in chemistry, biology, biotechnology and medicine.